Current Literature Describing 18F-FDG PET Texture Analysis in Response Prediction in EC
| Study | n | Type | nCRT | Timing PET/CT | Outcome | Reported entered variables |
| Tixier et al. (13) | 41 | AC, SCC | 60 Gy + cisplatin or carboplatin/fluorouracil | Pre-nCRT | CR vs. PR vs. NR (32) | Pre-local homogeneity, pre-local entropy, pre-coarseness, pre-intensity variability, pre-size-zone features |
| Hatt et al. (11) | 50 | AC, SCC | 60 Gy + cisplatin/fluorouracil | Pre-nCRT | CR + PR vs. NR (32) | Pre-MATV, pre-entropy, pre-homogeneity, pre-dissimilarity, pre-intensity variability, pre-zone percentage |
| Tan et al. (12) | 20 | AC, SCC | 50.4 Gy + cisplatin/fluorouracil | Pre- and post-nCRT | TRG 1 + 2 vs. 3–5 (19) | ΔSUVmax, SUVmax ratio, ΔSUVmean, pre-skewness, post-inertia, post-correlation, post-cluster prominence |
| Zhang et al. (15) | 20 | AC, SCC | 50.4 Gy + cisplatin or carboplatin | Pre- and post-nCRT | TRG 1 + 2 vs. 3–5 (19) | Post-orientation, tumor involves GEJ, ΔInertia, post-energy, post-entropy, Δskewness |
| Van Rossum et al. (14) | 217 | AC | 45 or 50.4 Gy + fluoropyrimidine with either a platinum compound or a taxane | Pre- and post-nCRT | TRG 1 vs. 2–4 (33) | EUS tumor length, cT stage, induction chemotherapy, post-nCRT endoscopic biopsy, subjective PET assessment, post-nCRT TLG, pre-cluster shade, Δrun percentage, ΔGLCM entropy, post-nCRT roundness |
| Current study | 97 | AC, SCC | 41.4 Gy + carboplatin/paclitaxel | Pre-nCRT | TRG 1 vs. 2–5 (19) | Histology, cT stage, pre-LRLGLe-PET, pre-RP-CT |
AC = adenocarcinoma; SCC = squamous cell carcinoma; CR = complete response; PR = partial response; NR = no response; MATV = metabolically active tumor volume; GEJ = gastroesophageal junction; EUS = endoscopic ultrasonography; cT = clinical T stage; TLG = total lesion glycolysis; GLCM = gray level co-occurrence matrix.